| PLEX86 | ||
Misuse of word "microcode" 102Misuse of word "microcode" 103 Morten Reistad My quick impression of "real microcode": (partly tongue in cheek) 1. It lives in expensive memory that is: - faster than main storage - wider than main storage - if possible a really strange number of... Misuse of word "microcode" 104 Jason Ozolins That was probably me. Microcode still makes sense as a way to achieve cross-cpu-model compatibility, but when somone... The Packard Bell 440 was later the Raytheon PB440, and extended to the 520. I loved it. The microcode ran out of regular system memory, although first there was small, faster (solid state-one cycle vs magnetic cores-two cycle) memory for as much of the microcode as would fit. But each major program had its own custom microcode interpreter. The Fortran compiler was written with one set of "buttembler" instructions, the fortran run time another set, and so on. At least some of the interpreted machines had a "DML" (decend to micro level) instruction, for including microcode in-line. I think the fortran run time took advantage of that when it would help. I ONLY programmed it in micro code, in hex, no compiler. The microcode was two 12 bit register to register or register-indirect instructions in one 24 bit word. Some interesting side effects: You could put two +127--128 relative jump instrictions in a word and jump twice as far. A test-and skip could skip either the second microinsruction in a word or the next whole word. Another reason to like the 440: all the boards were behind glbutt panels above the console desk, with lots blinking lights to view. The newer 520 was integrated circuids and all hidden down inside the console.
|
||||
Misuse of word "microcode" 103 Alt Folklore Computers from Newsgroups The #1 Usenet Provider on the Internet
|
||||